IT202100022505A1 - MANUFACTURING PROCEDURE OF A CAPACITIVE PRESSURE SENSOR AND CAPACITIVE PRESSURE SENSOR - Google Patents

Description

DESCRIPTION

of the patent for industrial invention entitled: ?MANUFACTURING PROCEDURE OF A CAPACITIVE PRESSURE SENSOR AND CAPACITIVE PRESSURE SENSOR?

The present invention ? relating to a method for manufacturing a microelectromechanical device (MEMS), and to a microelectromechanical device. In particular, the invention ? on the manufacture of a capacitive pressure sensor and the capacitive pressure sensor cos? obtained. The capacitive pressure sensor? equipped with a suspended region, or membrane, able to move with respect to the rest of the structure. In particular, this membrane represents a variable electrode, facing a fixed portion constituting a fixed electrode and separated from it by a cavity. partly or so buried.

Various techniques are known for manufacturing the membrane, based on the bonding of two substrates or on the removal of a sacrificial layer.

For example, US 6,521,965 provides for the manufacture of the lower electrode; providing a sacrificial region over the lower electrode; the epitaxial growth of the membrane layer; making attachment holes in the membrane layer; the removal of the sacrificial region through the attachment holes; and the closure of the holes by filling oxide. A similar process? also described by US 6,527,961 for making pressure sensors. US 6,012,336 uses silicon or metal nitride to fill the etching holes.

In the indicated processes, the filling of the attachment holes ? a critical stage. Indeed, isn't it? Is it possible to use a compliant material, otherwise this would penetrate into the cavity? just made and would lead to at least partial filling, with consequent false capacitive coupling. On the other hand, the use of a non-compliant material, also given the geometric characteristics of the holes, narrow and deep for applications in which ? a very thick membrane is required, it does not allow for complete closure. In fact, normally the attachment holes close in the vicinity? of the upper opening before the filling material has completely filled the holes in the lower part.

Even the use of two different materials, a first non-conforming material which narrows the upper opening and prevents a second conforming material from penetrating inside the cavity, does not solve the problem.

Purpose of the present invention? to provide a process and a device which overcome the drawbacks of the prior art.

According to the present invention, a method for manufacturing a MEMS device and a MEMS device so? obtained, as defined in the attached claims.

For a better understanding of the present invention, preferred embodiments thereof are now described, purely by way of non-limiting example, with reference to the attached drawings, in which:

figures 1-14 illustrate, in side sectional view, manufacturing steps of a microelectro-mechanical device, in particular a capacitive pressure sensor according to an embodiment of the present invention;

figure 15A illustrates, in side sectional view, a differential capacitive pressure sensor according to a further embodiment of the present invention;

- figure 15B illustrates a package in which the pressure sensor of figure 15A ? housed or houseable;

figure 16 illustrates a capacitive pressure sensor according to a further embodiment of the present invention; And

- figure 17 illustrates a capacitive pressure sensor according to a further embodiment of the present invention.

Figures 1-14 show successive steps of manufacturing a microelectromechanical device or system (MEMS) 30 according to an embodiment of the present invention. In particular, the MEMS 30 device integrates microelectromechanical structures for the transduction of one or more ambient pressure signals into respective electrical signals. In particular, the transduction ? performed on the basis of a variation of a capacity?. In the following, therefore, the MEMS device 30 is also referred to as a pressure sensor, or capacitive pressure sensor.

Figures 1-14 show a side sectional view of a wafer (?wafer?), in a triaxial reference system of axes X, Y, Z orthogonal to each other.

Figure 1 shows the wafer 1, having a front side 1a and a rear side 1b opposite each other along the axis Z, comprising a substrate 2 of semiconductor material, typically silicon. At the front side 1a, the substrate 2 is overlain by an insulating layer 3, for example silicon oxide (SiO2) having a thickness of between 0.2 and 2 ?m, typically 0.5 ?m. The insulating layer 3 ? for example formed by thermal oxidation.

In figure 2, a structural layer 4 is formed on the insulating layer 3, of electrically conductive material, for example of N-type doped polysilicon (e.g., with doping density between 1?10<19 > and 2?10< 20 >ions/cm<3>)). In one embodiment, the structural layer 4 is formed by deposition of polysilicon with LPCVD technique.

With reference to the pressure sensor of the capacitive type, the structural layer 4 forms a lower electrode of the pressure sensor (that is, the lower plate of the capacitor).

In the following, Figure 3, the structural layer 4 is shaped (eg, photolithographically) to define the shape desired and/or foreseen in the design phase of the lower electrode of the pressure sensor.

One then proceeds, Figure 4, with the formation, above the structural layer 4 (and above the exposed portions of the insulating layer 3) of an etch stop layer 5. 52 etching, according to one embodiment of the present invention, of aluminum oxide (Al2O3), also known as alumina. The attack interruption layer 5 has, for example, a thickness of a few tens of nanometers, for example comprised between 20 and 60 nm, in particular 40 nm.

Attack interrupt layer 5 ? formed by atomic layer deposition technique (ALD ? ?Atomic Layer Deposition?). The deposition of Al2O3 using the ALD technique? known in the state of the art and ? typically performed using trimethylaluminum (TMA, Al(CH3)3) and water vapor (H2O) as reagents. As an alternative to H2O vapours? possible to use ozone (O3). For example, the deposit can take place using TMA as the aluminum source and H2O as the oxidizer. Steven M. George's paper, Chem. Rev. 2010, 110, p. 111-131, or the document by Puurunen, R. L., J. Appl. Phys. 2005, 97, p. 121-301, describe possible methods of forming the attack interrupt layer 5.

The patent document WO 2013/061313 also discloses a method for forming an Al2O3 attack interrupter layer which can be used in the context of the present invention. In particular, as described in WO 2013/061313, the attack interrupt layer 5 is formed with a process which provides for the ALD deposition of two intermediate layers of Al2O3, both subjected to crystallization. The sequence of: i) deposition of a first intermediate layer of Al2O3, ii) crystallization of the first intermediate layer, iii) deposition of a second intermediate layer of Al2O3, and iv) crystallization of the second intermediate layer, allows the formation of a layer of attack interruption 5, of Al2O3 with characteristics of resistance to attack by solutions containing hydrofluoric acid (HF) and, above all, impermeability? of the attack interrupt layer 5 to such HF-based solutions.

Furthermore, this attack interruption layer 5, in addition to being resistant to attack with HF and impermeable to HF, shows excellent properties of adhesion to the underlying silicon oxide layer 3 and to the polysilicon layer 4, shows excellent properties? dielectrics which do not vary as a function of any subsequent heat treatments, shows little (negligible) variation of the radius of curvature (?warpage?) of the wafer 1, and shows a high compatibility? with high temperature thermal processes (above 1000?C).

Then, figure 5, a sacrificial layer 8 is formed, for example of silicon oxide. The thickness of this sacrificial layer 8, at and above the structural layer 4, is between 0.4 and 2 ?m (or in any case chosen according to the capacity value?). To make up for the presence of the ?step? between the insulating layer 3 and the structural layer 4, and forming a sacrificial layer 8 having a planar upper surface, a planarization step is performed after the formation of the sacrificial layer 8 (e.g., by means of CMP).

Alternatively, ? It is possible to form the sacrificial layer 8 in two successive sub-steps, comprising:

- forming a first sacrificial substrate 8a, here of silicon oxide deposited with the PECVD technique (TEOS or silane-based oxide), up to complete coverage of the attack interrupting layer 5 in the region of the same which extends above the structural layer 4 ; the thickness tox1 of the first sacrificial substrate 8a, measured along the Z axis laterally to the structural layer 4, is? greater than the sum of the thicknesses of the structural layer 4 and of the attack interruption layer 5 (e.g. between 700 nm and 1.5 ?m);

- planarizing the first sacrificial substrate 8a, for example with the CMP technique, so as to obtain a top surface of the same planar but without exposing portions of the underlying attack interrupting layer 5;

- forming a second sacrificial substrate 8b, here of silicon oxide deposited with the PECVD technique (TEOS or silane-based oxide), above the first sacrificial substrate 8a; the thickness tox-c of the second sacrificial substrate 8b, measured along the Z axis starting from the upper surface of the first sacrificial substrate 8a, is? between 300 nm and 2 ?m.

The first and second sacrificial substrate 8a, 8b together form the sacrificial layer 8. The thickness of this sacrificial layer 8, ? chosen according to the value of the capacity? desired (e.g., between 500 nm and 2.3 ?m).

Then, Figure 6, is an etching of the sacrificial layer 8 carried out so as to form a trench 10 which surrounds, or internally delimits, a region 8? of the sacrificial layer 8. The trench 10 extends along the Z axis for the entire thickness of the sacrificial layer 8. In this way, the region 8? ? separated from the remaining portions of the sacrificial layer 8 by the trench 10. The shape of the region 82, defined by the trench 10, corresponds to the desired shape of the cavity? through which the two conductive plates of the condenser which form the sensitive element of the pressure sensor face each other, as better evident from the continuation of the description.

Then, Figure 7, one proceeds with the formation, above the sacrificial layer 8 (including the region 8?) and in the trench 10, of a further attack interruption layer 15. The attack interruption layer 15?, according to a form embodiment of the present invention, of aluminum oxide (Al2O3), with a thickness of a few tens of nanometres, for example comprised between 20 and 60 nm, in particular 40 nm.

The etch interrupt layer 15 is formed according to the same process previously discussed for the etch interrupt layer 5.

Then, figure 8, the attack interrupt layer 15 is shaped (?patterned?) by removing selective portions of the same above the region 8?, forming a cavity 15a through the attack interrupt layer 15, until it reaches the surface of the region 8?. At least a portion of the surface of region 8? ? what? exposed through the cavity? 15a. The region 8 area? exposed in this process step (that is, the cavity 15a) defines the shape and the spatial extension of the upper plate of the condenser which forms the active element of the pressure sensor, as better evident from the continuation of the description.

Then, figure 9, a deposition step of a structural layer 16 is performed above the attack interruption layer 15 and in the cavity? 15a, covering the surface of region 8?. In one embodiment, the structural layer 16 is of conductive material, for example of doped polysilicon (e.g., with doping between 1?10<18 > and 2?10<20 >ions/cm<3>). Alternatively, the structural layer 16 can be of undoped polysilicon.

The structural layer 16 ? for example deposited with the LPCVD technique. The structural layer 16 has a thickness, for example, of between 0.2 ?m and 1 ?m.

In the following, figure 10, the structural layer 16 is defined, for example photolithographically, to selectively remove it at the cavity 15a. In particular, in one embodiment, the structural layer 16 is not completely removed at the cavity? 15a, so as to leave regions 16? which act as an anchor for a subsequent layer that will come? deposited later (layer 20, illustrated in Figure 11). It is clear that, in other embodiments, if it is believed that this anchoring is not necessary to structurally support the layer 20 of figure 11, the regions 16? are not formed and the structural layer 16 is completely removed at the cavity? 15a.

Therefore, Figure 11, as anticipated, is a permeable layer 20 formed above the structural layer 16, of the anchoring regions 16? (if any) and region 8? exposed between the anchoring regions 16?.

The permeable layer 20?, in one embodiment of the present invention, of polysilicon permeable to the chemical solution used for subsequent removal of the region 8?. For example, in the disclosed embodiment, wherein region 8? ? of silicon oxide, ? You can use hydrofluoric acid (HF), or solutions containing HF, to selectively remove the 8? region. In this case, the permeable layer 20 ? provided with pores or openings able to allow the passage of the hydrofluoric acid through the permeable layer 20, reaching and removing the region 8? and forming a cavity? or buried chamber 22.

The permeable layer 20 ? in particular of polycrystalline silicon, having holes (or pores) with a diameter ranging from 1 to 50 nm. The thickness of the permeable layer 20 ? in the range of 50 to 150 nm, for example 100 nm. The permeable layer 20 ? for example deposited using the LPCVD technique. According to an exemplary, non-limiting embodiment, the deposition conditions are in the tensile-to-compression transition region, with a process window around 600°C using a silane source gas, in a deposition environment with a pressure of about 550 mtorr. The pore sizes of the permeable layer 20 are, in general, chosen such that the etching solution (liquid or gaseous) used to remove the region 8? can penetrate through the pores until it reaches the permeable layer 20.

In general, the permeable layer 20 can be porous polysilicon, formed in a manner known in the literature, or polysilicon having holes (apertures) actively formed following its deposition, by mechanical or physicochemical action of selective removal of material.

Referring to Figure 12 , an etching step of region 8? (identified by arrows 21) is performed with HF or blends of buffered HF or with vapor etch techniques using HF in vapor form. Region 8 material? is completely removed and the cavity is formed? buried 22. As mentioned, the chemical agent used for the attack permeates through the openings or pores of the permeable layer 20.

Then, Fig. 13 , a sealing layer 24 is formed (e.g., by performing an epitaxial growth of amorphous silicon) on the permeable layer 20, to form a second electrode operatively coupled to the first electrode (i.e., the layer 4 formed in Fig. 3 ) through the cavity? 22. The sealing layer 24 has a thickness for example between 0.2 ?m and 2 ?m. The amorphous silicon of the sealing layer 24 can? be deposited with the PECVD technique, at a deposition temperature between 200 and 400 ?C, using SiH4/H2 or SiH4/He as precursors. If required by the application, the sealing layer 24 can be doped in situ using phosphine (PH3) or biborane (B2H6). In the context of the present invention, the sealing layer is electrically conductive (made so by doping).

One or more? further layers 25 can be deposited or formed on the sealing layer 24, for example one or more? layers of a respective material among among: polysilicon, Al2O3,HfO2, SiN (PE) with passivation or reinforcement function.

If necessary, the sealing of the permeable layer 20 (and therefore of the buried cavity 22) can take place in an environment (reaction chamber) with controlled pressure, in order to set a desired pressure value in the cavity? buried 22. This pressure value can? vary for example between 0.09 mbar and 205 mbar.

Alternatively, however, it is noted that the use of the PECVD technique to form the sealing layer 24, by depositing amorphous silicon, allows the generation of a desired pressure in the cavity? buried layer 22. In fact, the product between the deposition temperature of layer 24 (about 350? C) and the pressure at which one works in the reaction chamber (about 1.5 Torr) allows to have, once layer 24 has cooled , a low pressure inside the cavity? 22.

Then, with reference to Figure 14, conductive pads 28, 29 are formed to allow the sensing electrodes of the pressure sensor to be biased from the outside so as to manufactured. A conductive pad 28 ? electrically coupled to the sealing layer 24, while the other conductive pad 29 is? electrically coupled to the structural layer 4, laterally to the cavity? buried 22. The pads 28, 29 are formed by depositing conductive material, for example metal such as aluminum, and shaping it to achieve the desired extents for the pads.

In one embodiment, to place the conductive pad 29 in contact with the structural layer 4, a selective removal step of the layers 25, 24, 20, 16, 15 and possibly of the layer 8 (if present in the area where you want to form pad 29).

In order to protect the layers exposed through the opening so? formed, the formation of one or more? layers 25 previously described pu? take place after the formation of such an opening (therefore one or more layers 25 is also deposited inside this opening) and before the formation of the pad 29.

Alternatively, the conductive pad 29 can be brought into electrical contact with the structural layer 4 by a conductive path extending between the structural layer 4 and the conductive pad 29.

Is it formed like this? a MEMS device 30, in particular a pressure sensor of the capacitive type, even more? in particular, an absolute pressure sensor configured to detect a pressure variation outside the sensor with respect to the pressure value present inside the cavity? buried 22 (fixed value, set, as described, during manufacturing).

The pressure sensor 30 ? provided with a support body (substrate 2 plus layer 3) on which extends the first electrode (layer 4) of the capacitor used for capacitive sensing. The first electrode? facing the cavity? buried 22 (in particular, with the interposition of layer 5). Above the cavity buried 22, opposite the first electrode 4, extends the second electrode (layer 20 plus layer 24). Are the first and second electrodes therefore facing each other through the cavity? buried 22. The second electrode ? a membrane configured to deflect along the Z axis. The pressure variation in the environment external to the pressure sensor 30 causes a deflection of the second electrode and a consequent variation in capacitance? of the capacitor cos? format, which is detected by means of the conductive pads 28, 29, so in itself? known and processed by known circuitry, not shown.

According to a different and further embodiment (figure 15A), the MEMS device ? a pressure sensor of the differential capacitive type 30?, configured to supply an identification signal of the difference between two ambient pressures at which the sensor itself? exposed. The pressure sensor 30? ? manufactured according to the same steps previously described for the pressure sensor 30 (figures 1-14), with the exception of the cavity? 22 which must be connected to the outside in order to be able to operate the sensor 30? as a differential sensor. To this end, the cavity 22 ? fluidically connected to the outside of the pressure sensor 30?, for example through a suitably made channel which allows the passage of air (or other fluid in gaseous form) towards the cavity? 22. The resulting deformation of the membrane (second electrode) ? indicative of the difference between a first ambient pressure P1 (outside the cavity 22) and a second ambient pressure P2 (inside the cavity 22), and the signal transduced by the differential pressure sensor 30? ? a differential pressure signal.

With reference to figure 15B, the differential pressure sensor 30? ? provided with a package 32 (common elements of the pressure sensor 30? with the pressure sensor 30 are identified with the same reference numbers). Package 32 includes an internal housing 33 in which ? housed or arranged the differential pressure sensor 30?. The package 32 has a first through opening 32a, configured to put the membrane (second electrode) of the differential pressure sensor 30 into fluid communication. with the environment external to the package 32, and to form an access channel of the pressure P1 towards the membrane (but not towards the cavity 22). The package 32 also has a second through opening 32b, configured to put the cavity in fluid communication 22 with the? external environment to the package 32, and to form a channel of access of the pressure P2 towards the cavity? 22 (but not towards the membrane). Are the first and second through openings 32a, 32b therefore formed and connected to the pressure sensor 30? in such a way that, inside the package 32, the pressures P1 and P2 remain separate, in order to allow the correct functioning of the pressure sensor in modality? differential. In other words, the pressure sensor 30? ? mounted in the package 32 in such a way that the access channel to the cavity? 22 is connected to the second through opening 32b by means of suitable fluid tight means or systems (watertight), preventing a fluidic connection of the second through opening 32b with other regions of the internal chamber of the package 32.

The differential pressure sensor 30? therefore lends itself to being mounted in systems / components in which the first through opening 32a ? in communication with a first room at ambient pressure P1 and the second through opening 32b ? in communication with a second environment at ambient pressure P2. The first through opening 32a therefore forms an access for the pressure P1 which acts on a first side of the membrane (e.g., the side outside the cavity 22), deforming it. The second through opening 32b forms a respective access for the pressure P2 which acts on a second side, opposite to the first side (e.g., internal side of the cavity 22), of the membrane, generating a deformation force of the membrane which counteracts the force generated from the pressure P1. The resulting deformation of the membrane ? indicative of the difference between the pressure P1 and the pressure P2, and the signal transduced by the differential pressure sensor 30? ? a differential pressure signal.

The patent documents US7,763,487 and US8,008,738 disclose packages usable in the context of the present invention, to encapsulate a pressure sensor 30? differential type.

Figure 16 illustrates a further embodiment of a MEMS device according to the present invention, applicable both to the pressure sensor 30 and to the pressure sensor 30?. The MEMS device shown in figure 16 comprises all the elements and technical characteristics previously described in the respective embodiments.

The MEMS device of figure 16 also comprises a further cavity? or buried chamber 42 extending into the substrate 2, below the first electrode (ie below the layer 4). This cavity? buried 42 extends, for example, to a distance d1 (measured along the Z axis) from the bottom of the cavity? 22 between 20 ?m and 60 ?m. In this way, the portion of slice 1 above the cavity? buried 42 forms a further membrane that pu? deflect to discharge any residual stress from manufacturing or that may arise during use of the MEMS 30, 30? device, preventing any structural problems such as breaks, cracks, deformations.

The cavity? buried 42 pu? be formed in a way per s? known, for example according to the process of formation of cavities? buried described in US7,763,487 and US8,008,738.

According to a further embodiment of the present invention, illustrated in figure 17, both the MEMS device 30 and the MEMS device 30? do they include a respective layer of anti-adhesion (?anti-stiction?) 50 inside the cavity? buried 22 and/or of the cavity? buried 42 (if present). For simplicity? of description and illustration, figure 17 does not show the cavity? buried 42, however, as mentioned above, what has been described also applies to the case in which this cavity buried 42 is present.

The anti-adhesion layer 50 can? completely cover the internal walls of the cavity? 22 (and/or cavity 42), or only partially.

The anti-adhesion layer 50 ? of a material chosen in such a way as to limit or avoid even partial occlusion of the cavity? 22 (and/or cavity 42) due to a potential reciprocal adhesion of the walls which delimit the cavity at the top and bottom 22 (and/or cavity 42). Would this undesirable effect cause the impossibility? of correct movement of the second electrode and consequent failure of the MEMS 30/30? device.

To this end, the anti-adhesion layer 50 can be introduced into the cavity? 22 through a suitable opening that connects the cavity? 22 with the environment in which the deposition of the anti-adhesion layer 50 takes place. This opening can? later be closed in the case of the absolute pressure sensor 30, or can? be the opening used to put the cavity in fluidic communication? 22 with the external environment in the case of differential pressure sensor 30?, which therefore remains fluidically accessible.

The deposition of the anti-adhesion layer 50 can? take place by means of a vapor phase process.

Usable materials for the anti-adhesion layer 50 include, but are not limited to, chlorosilanes, trichlorosilanes, dichlorosilanes, siloxanes, etc., such as:

DDMS ? ?dimethyldichlorosilane?;

FOTS? ?perfluorooctyltrichlorosilane?;

PF10TAS ? ?perfluorodecyltris(dimethylamino)silane?; PFDA ? ?perfluorodecanoic acid?.

Usable materials, and the related deposition processes, are known in the state of the art, in particular from Ashurst, W. & Carraro, C. & Maboudian, Roya. (2004), ?Vapor phase anti-stiction coatings for MEMS? Device and Materials Reliability, IEEE Transactions on. 3. 173-178. 10.1109/TDMR.2003.821540.

The manufacturing processes and devices described above, according to the various embodiments, have numerous advantages.

Thanks to the monolithic structure of the membrane, substantially free of empty areas, the membrane is robust and therefore particularly suitable for the construction of MEMS structures of different types, reducing the risk of breakage, deformation or damage that could compromise its functionality.

The process ? of simple realization, given that it does not present particular criticalities? or difficulty? of execution, thus guaranteeing high yields and reduced final costs. It is also noted that the manufacturing process requires the use of a single wafer of semiconductor material, resulting in economically advantageous and with reduced criticality? due to the absence of gluing or bonding phases between slices.

Furthermore, the manufacturing process ? particularly flexible, as it allows you to create cavities? buried and/or membranes of the desired shape and size, both in terms of area and thickness, in a simple way. In particular, for the application as a pressure sensor, ? It is possible to obtain a high thickness of the membrane, in order to increase the accuracy of the sensor itself.

The use of porous silicon guarantees the obtainment of a membrane with a regular shape and avoiding unwanted formations which would compromise or in any case reduce the electrical/mechanical characteristics of the finished MEMS device.

The simultaneous presence of the two layers of crystallized aluminum oxide prevents short circuits between the upper and lower electrodes of the capacitor and allows the diameter of the membrane to be defined during the manufacturing phase, which is not dependent on the attack time.

Furthermore, thanks to the use of the two layers of crystallized aluminum oxide, ? It is possible to precisely define the size of the membrane, without using a time attack. In fact, the crystallized aluminum oxide acts as a hard mask for the subsequent HF attack aimed at removing the oxide layer under the membrane.

The use of an HF permeable polysilicon layer enables the formation of a porous grid which allows the HF to permeate and attack the oxide. The permeable polysilicon also serves as a support for the upper layers.

The use of the amorphous silicon layer 24 (deposited via PECVD), thanks to which the closure of the porous polysilicon layer? fast, also allows the definition of a desired pressure in the cavity? buried layer 22. In fact, the product between the deposition temperature (about 350? C) and the pressure at which one works in the reaction chamber (about 1.5 Torr) allows to have, once the amorphous silicon layer 24 has cooled , a vacuum pushed inside the cavity? 22. Furthermore, using amorphous silicon, and performing a PECVD deposition, the volume of the cavity? 22 is not reduced by unwanted or waste products.

Finally, it is clear that modifications and variations can be made to the process and device described and illustrated here without thereby departing from the protective scope of the present invention, as defined in the attached claims.

The teaching of the present invention can be used to make MEMS devices of a different type than those described, such as accelerometers, gyroscopes, resonators, valves, inkjet print heads and the like, in which case the structures below and/or above the membrane are adapted according to the intended application. In each case, the size, shape and number of channels are optimized according to the application and the MEMS device is completed with the structures and elements necessary for its operation.

In the event that it is desired to integrate electronic components in the same slice 1, this can? be carried out using the substrate 2 or further epitaxial layers formed on top of the sealing layer 24.

Claims (22)

1. Process for manufacturing a micro-electro-mechanical device (30; 30?), comprising the steps of: forming, on a substrate (2), a first protective layer (5) impermeable to a chemical etching solution; forming, on the first protection layer (5), a sacrificial layer (8, 8?) of a material that can? be removed by said chemical etch solution; forming, on the sacrificial layer (8, 8?), a second protection layer (15) impermeable to said chemical etching solution; selectively removing a portion of the second protective layer (15) to expose a respective sacrificial portion (8?) of the sacrificial layer (8, 8?); forming, on said sacrificial portion (8?), a first membrane layer (20) of a porous material, which ? permeable to said etching chemistry; form a cavity (22) removing the sacrificial portion (8?) through the first membrane layer (20) using said etching chemistry; And sealing pores of the first membrane layer (20) by forming a second membrane layer (24) on the first membrane layer (20). 2. Process according to claim 1, wherein said chemical etching solution comprises hydrofluoric acid, HF, and said first and second protective layers (5, 15) include crystallized Aluminum Oxide. The method according to claim 1 or claim 2, wherein the first membrane layer (20) is of porous silicon or silicon presenting a plurality? of through holes or pores. 4. Process according to any one of the preceding claims, wherein forming the second membrane layer (24) comprises depositing doped amorphous silicon by means of the PECVD technique. 5. Process according to any one of the preceding claims, further comprising the step of forming, on the substrate (2), a conductive layer (4), wherein the step of forming the first protection layer (5) comprises forming the first protection layer (5) on top of the conductive layer (4), said conductive layer (4) and said second membrane layer (24) being capacitively coupled together through the cavity? (22). 6. Method according to claim 5, wherein the conductive layer (4) is of doped polysilicon. 7. Process according to any one of the preceding claims, further comprising the step of forming a buried chamber (42) in the substrate (2) below the, and at least partially aligned with, the cavity? (22). 8. Process according to any one of the preceding claims, further comprising the step of fluidically connecting the cavity? (22) with an environment external to said micro-electro-mechanical device (30; 30?) via a through opening. 9. Process according to claim 8, further comprising the step of internally covering the cavity? (22) by means of an anti-adhesion layer by causing chemical species including chlorosilanes, trichlorosilanes, dichlorosilanes, siloxanes to flow through said through opening. 10. Process according to claim 8 or 9, further comprising the steps of: disposing said micro-electro-mechanical device (30; 30?) inside a package (32) having an internal housing and being provided with a first access channel (32a; 32b) and a second access channel (32b 32a) toward said inner housing; And coupling said one of the first and second access channels (32a; 32b) to the through opening by fluid tight means or systems configured to prevent a fluidic connection between said inner package housing (32) and said cavity? (22). 11. Process according to any one of the preceding claims, wherein said microelectro-mechanical device (30; 30?) ? a capacitive pressure sensor. 12. Micro-electro-mechanical device (30; 30?), comprising: a substrate (2); a first protective layer (5) impermeable to a chemical etching solution, extending over the substrate (2); a sacrificial layer (8, 8?) of a material that pu? be removed by means of said chemical etching solution, extending over the first protective layer (5); a second protective layer (15) impermeable to said chemical etching solution, extending over the sacrificial layer (8, 8?); a first membrane layer (20) of a porous material which is permeable to said etching chemistry; a cavity? (22) extending between the first membrane layer (20) and the first protection layer (5); And a second membrane layer (24) on the first membrane layer (20), configured to seal pores of the first membrane layer (20). The device according to claim 12, wherein said chemical etching solution comprises hydrofluoric acid, HF, and said first and second protective layers (5, 15) include crystallized Aluminum Oxide. The device according to claim 12 or claim 13, wherein the first membrane layer (20) is? of porous silicon or silicon presenting a plurality? of through holes or pores. 15. Device according to any one of claims 12-14, wherein the second membrane layer (24) is? of doped amorphous silicon. The device according to any one of claims 12-15, further comprising a conductive layer (4) on the substrate (2), wherein the first protective layer (5) extends above the conductive layer (4), and wherein said conductive layer (4) and said second membrane layer (24) are capacitively coupled together through the cavity? (22). 17. Device according to claim 16, wherein the conductive layer (4) is of doped polysilicon. 18. A device according to any one of claims 12-17, further comprising a chamber (42) buried in the substrate (2) below and at least partially aligned with the cavity? (22). 19. Device according to any one of claims 12-18, further comprising a fluidically connected through opening between the cavity? (22) and an environment external to said micro-electro-mechanical device (30; 30?). 20. A device according to claim 19, further comprising an anti-adhesion layer internally covering the cavity? (22), the anti- layer including a material among chlorosilane, trichlorosilane, dichlorosilane, siloxane. 21. Device according to claim 19 or 20, further comprising: a package (32) having an internal housing and being provided with a first access channel (32a; 32b) and with a second access channel (32b; 32a) towards said internal housing, wherein one of the first and second access channels (32a; 32b) is coupled to the through opening by fluid tight means or systems configured to prevent a fluid connection between said inner package housing (32) and said internal cavity (32). (22). 22. Device according to any one of the preceding claims, wherein said microelectro-mechanical device (30; 30?) ? a capacitive pressure sensor.